docs/transformers/tests/models/bridgetower/test_processor_bridgetower.py

# Copyright 2023 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import shutil
import tempfile
import unittest

from transformers.testing_utils import require_torch, require_vision
from transformers.utils import is_vision_available

from ...test_processing_common import ProcessorTesterMixin


if is_vision_available():
    from transformers import (
        AutoProcessor,
        BridgeTowerImageProcessor,
        BridgeTowerProcessor,
        RobertaTokenizerFast,
    )


@require_vision
class BridgeTowerProcessorTest(ProcessorTesterMixin, unittest.TestCase):
    processor_class = BridgeTowerProcessor

    @classmethod
    def setUpClass(cls):
        cls.tmpdirname = tempfile.mkdtemp()

        image_processor = BridgeTowerImageProcessor()
        tokenizer = RobertaTokenizerFast.from_pretrained("BridgeTower/bridgetower-large-itm-mlm-itc")

        processor = BridgeTowerProcessor(image_processor, tokenizer)

        processor.save_pretrained(cls.tmpdirname)

    def get_tokenizer(self, **kwargs):
        return AutoProcessor.from_pretrained(self.tmpdirname, **kwargs).tokenizer

    def get_image_processor(self, **kwargs):
        return AutoProcessor.from_pretrained(self.tmpdirname, **kwargs).image_processor

    @classmethod
    def tearDownClass(cls):
        shutil.rmtree(cls.tmpdirname, ignore_errors=True)

    # Some kwargs tests are overriden from common tests to handle shortest_edge
    # and size_divisor behaviour

    @require_torch
    @require_vision
    def test_image_processor_defaults_preserved_by_image_kwargs(self):
        if "image_processor" not in self.processor_class.attributes:
            self.skipTest(f"image_processor attribute not present in {self.processor_class}")
        image_processor = self.get_component(
            "image_processor",
            crop_size={"shortest_edge": 234, "longest_edge": 234},
        )
        tokenizer = self.get_component("tokenizer", max_length=117, padding="max_length")

        processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
        self.skip_processor_without_typed_kwargs(processor)

        input_str = "lower newer"
        image_input = self.prepare_image_inputs()

        inputs = processor(text=input_str, images=image_input)
        self.assertEqual(len(inputs["pixel_values"][0][0]), 234)

    @require_torch
    @require_vision
    def test_structured_kwargs_nested_from_dict(self):
        if "image_processor" not in self.processor_class.attributes:
            self.skipTest(f"image_processor attribute not present in {self.processor_class}")

        image_processor = self.get_component("image_processor")
        tokenizer = self.get_component("tokenizer")

        processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
        self.skip_processor_without_typed_kwargs(processor)
        input_str = "lower newer"
        image_input = self.prepare_image_inputs()

        # Define the kwargs for each modality
        all_kwargs = {
            "common_kwargs": {"return_tensors": "pt"},
            "images_kwargs": {
                "crop_size": {"shortest_edge": 214},
            },
            "text_kwargs": {"padding": "max_length", "max_length": 76},
        }

        inputs = processor(text=input_str, images=image_input, **all_kwargs)
        self.assertEqual(inputs["pixel_values"].shape[2], 214)

        self.assertEqual(len(inputs["input_ids"][0]), 76)

    @require_torch
    @require_vision
    def test_kwargs_overrides_default_image_processor_kwargs(self):
        if "image_processor" not in self.processor_class.attributes:
            self.skipTest(f"image_processor attribute not present in {self.processor_class}")
        image_processor = self.get_component("image_processor", crop_size={"shortest_edge": 234})
        tokenizer = self.get_component("tokenizer", max_length=117)
        if not tokenizer.pad_token:
            tokenizer.pad_token = "[TEST_PAD]"
        processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
        self.skip_processor_without_typed_kwargs(processor)

        input_str = "lower newer"
        image_input = self.prepare_image_inputs()
        inputs = processor(text=input_str, images=image_input, crop_size={"shortest_edge": 224})
        self.assertEqual(len(inputs["pixel_values"][0][0]), 224)

    @require_torch
    @require_vision
    def test_unstructured_kwargs_batched(self):
        if "image_processor" not in self.processor_class.attributes:
            self.skipTest(f"image_processor attribute not present in {self.processor_class}")
        image_processor = self.get_component("image_processor")
        tokenizer = self.get_component("tokenizer")
        if not tokenizer.pad_token:
            tokenizer.pad_token = "[TEST_PAD]"
        processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
        self.skip_processor_without_typed_kwargs(processor)

        input_str = ["lower newer", "upper older longer string"]
        image_input = self.prepare_image_inputs(batch_size=2)
        inputs = processor(
            text=input_str,
            images=image_input,
            return_tensors="pt",
            crop_size={"shortest_edge": 214},
            padding="longest",
            max_length=76,
        )
        self.assertEqual(inputs["pixel_values"].shape[2], 214)

        self.assertEqual(len(inputs["input_ids"][0]), 6)

    @require_torch
    @require_vision
    def test_unstructured_kwargs(self):
        if "image_processor" not in self.processor_class.attributes:
            self.skipTest(f"image_processor attribute not present in {self.processor_class}")
        image_processor = self.get_component("image_processor")
        tokenizer = self.get_component("tokenizer")
        if not tokenizer.pad_token:
            tokenizer.pad_token = "[TEST_PAD]"
        processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
        self.skip_processor_without_typed_kwargs(processor)

        input_str = "lower newer"
        image_input = self.prepare_image_inputs()
        inputs = processor(
            text=input_str,
            images=image_input,
            return_tensors="pt",
            crop_size={"shortest_edge": 214},
            padding="max_length",
            max_length=76,
        )

        self.assertEqual(inputs["pixel_values"].shape[2], 214)
        self.assertEqual(len(inputs["input_ids"][0]), 76)

    @require_torch
    @require_vision
    def test_structured_kwargs_nested(self):
        if "image_processor" not in self.processor_class.attributes:
            self.skipTest(f"image_processor attribute not present in {self.processor_class}")
        image_processor = self.get_component("image_processor")
        tokenizer = self.get_component("tokenizer")
        if not tokenizer.pad_token:
            tokenizer.pad_token = "[TEST_PAD]"
        processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
        self.skip_processor_without_typed_kwargs(processor)

        input_str = "lower newer"
        image_input = self.prepare_image_inputs()

        # Define the kwargs for each modality
        all_kwargs = {
            "common_kwargs": {"return_tensors": "pt"},
            "images_kwargs": {"crop_size": {"shortest_edge": 214}},
            "text_kwargs": {"padding": "max_length", "max_length": 76},
        }

        inputs = processor(text=input_str, images=image_input, **all_kwargs)
        self.skip_processor_without_typed_kwargs(processor)

        self.assertEqual(inputs["pixel_values"].shape[2], 214)

        self.assertEqual(len(inputs["input_ids"][0]), 76)